This Selected Issues paper examines economic growth in Switzerland. It attempts to analyze whether slow growth is inescapable for Switzerland. The paper suggests that income convergence across countries contributes significantly to slow relative growth in Switzerland, but experience in several advanced industrial countries reinforces the view that slow growth is not inescapable. Higher growth will require raising total factor productivity growth, which remains low by international standards, and to a lesser extent, raising the investment rate.

Abstract

This Selected Issues paper examines economic growth in Switzerland. It attempts to analyze whether slow growth is inescapable for Switzerland. The paper suggests that income convergence across countries contributes significantly to slow relative growth in Switzerland, but experience in several advanced industrial countries reinforces the view that slow growth is not inescapable. Higher growth will require raising total factor productivity growth, which remains low by international standards, and to a lesser extent, raising the investment rate.

II. Direction for Further Reform of the Swiss Tax and Revenue System12

This note argues that Switzerland’s tax burden is not as low as appears at first sight. By international standards, its structure appears skewed toward the taxation of capital. Rebalancing the system toward the taxation of consumption, and removing distortions within the structure of capital taxation, could yield efficiency gains for the economy.

A. Switzerland’s Tax Burden

28. By the standards of industrialized countries, Switzerland is blessed with a relatively low tax burden. This adds to its attractiveness as a place to do business and means that distortions to economic decisions, and the associated dead-weight losses, remain contained. Among 20 highly-advanced OECD countries, Switzerland’s tax ratio amounts to 36.2 percent of GDP against 38.8 percent of GDP on average (Figure II-1). Likewise, total general government revenue equivalent to 37.3 percent of GDP remains well below the average of 43.8 percent of GDP (Figure II-2).

Figure II-1.
Figure II-1.

Selected OECD Countries: General Government Tax Revenue, 2000 1/

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: OECD Revenue Statistics: and Analytical Database.1/ Data For Australia and the United States relate to 1999.
Figure II-2.
Figure II-2.

Selected OECD Countries: General Government Revenue, 2000

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: IMF World Ecomomic Outlook.

29. Switzerland’s edge over its peers reflects to a large extent a narrower coverage of social security in the revenue statistics, however. Unlike many other countries, Switzerland collects the majority of pension contributions under a privately-operated second pillar of the system. Such contributions are not captured by government revenue statistics, although they are government-mandated. This treatment is consistent with the usual delineation of the public and private sectors. But it also means that statistics record less than half of pension contributions in Switzerland while capturing almost all pension contributions in countries (e.g., Germany) that predominantly rely on a semi-publicly-operated pension system. A similar coverage issue concerns health care contributions. Mandatory premiums to privately-operated insurance, which are sizable in Switzerland, are not considered government revenue. But contributions to the semi-publicly-operated sickness funds in Germany are. These discrepancies in coverage explain more than half the difference between the revenue take in Switzerland and Germany (Table II-1).13

Table II-1.

Switzerland, Germany: Government Revenue, 1994-2000

(In percent of GDP)

article image
Sources: Eidgenössische Finanzverwaltung; Bundesamt für Sozialversicherung; Bundesfinanzministerium; Verband der privaten Krankenversicherer; and IMF staff estimates.

Fifty percent of contributions to the second pillar are classified as mandatory here, in line with government estimates.

Private health insurance coverage is mandatory for those opting out of the public health care insurance.

30. Revenue including compulsory contributions might be a better indicator for the government-imposed financial burden and the associated distortions. From an economic point of view, government-mandated contributions to private-sector funds have much the same effect as social security contributions to a semi-publicly-operated system. Distortions to labor supply decisions, for example, arise mainly from the wedge between compensation and take-home pay. They might be mitigated to the extent that pension contributions are closely linked to future benefits. But it is the redistribute nature of the entire pension system, rather than the form of the deductions per se, that is critical in this context. In Switzerland, benefits under the second pillar are purely earnings related but benefits under the first pillar are highly redistributive (Queisser and Vittas, 2000). In Germany, the first pillar is less redistributive but makes up a much larger share of the pension system.

31. Regardless as to whether mandatory contributions imply a higher tax burden, Switzerland’s pension system still conveys considerable benefits over a PAYG system. The fact that its second pillar is large and fully funded means that Switzerland is much better prepared to weather the strains from demographic aging than most advanced economies. Moreover, public debt is relatively low, just over 50 percent of GDP, and pension fund assets are substantial (Figure II-3), amounting to some 120 percent of GDP (Table II-2).

Figure II-3.
Figure II-3.

Selected OECD Countries: General Government Debt,2001

(In percent of GDP)

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: IMF World Economic Outlook.
Table II-2.

Switzerland: Capitalization of Second Pillar Pension Funds, 1998-2001

(In percent of GDP)

article image
Sources: Eidgcnossische Finanzvervallung; Bundesamt filr Sozialversicherung; and SBB.

Dissection according to legal form of pension fund.

B. Switzerland’s Tax Structure

32. The Swiss tax system places a relatively high tax burden on capital income while taxing consumption and labor lightly. The tax burden is commonly measured by average effective tax rates (AETRs), calculated as the ratio of actual tax collection from a factor divided by an appropriate measure of the tax base. In the case of capital income, it expresses revenue collections under the corporate income tax, property taxes, capital gains taxes, and that part of the income tax that relates to household capital income, as a fraction of the net operating surplus of the economy. Calculations by Carey and Tchilinguirian (2000) put Switzerland among the most heavy taxers of capital and the lightest taxers of labor and consumption (Figure II-4 and Figure II-5).

Figure II-4.
Figure II-4.

Selected OECD Countries: AETR on Capital, 1991-97

(in percent)

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: Cary and Tchilinguirian (2000).
Figure II-5.
Figure II-5.

Selected OECD Countries: AETR on Labor and Consumption Combined, 1991-97

(In percent)

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: Carey and Tchilinguirian (2000).

33. The relatively high burden on capital income appears not to be driven by taxes on corporations. Indeed, corporation taxes account for a smaller share in the AETR on capital than in the average comparator country (Figure II-6). On the other hand, taxes on interest/dividends/profits of unincorporated businesses and property taxes make a high contribution by international standards. This likely reflects a quite rigorous withholding tax regime on interest income and dividends and the prevalence of net wealth taxes and transaction taxes.

Figure II-6.
Figure II-6.

Selected OECD Countries: Contributions to AETR on Capital by Tax, 1991-1997

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: IMF staff estinates bascd on the dataset of carey and Tchilinguirian (2000).

34. Although the AETR methodology has its drawbacks, the finding of a high relative capital tax burden appears to be a fairly robust result. One problem is the dissection of income tax collections into capital and labor taxes, a breakdown that revenue statistics rarely provide. The AETR-methodology assumes that both sources of income are taxed at the same rate, as under an ideal global income tax. But real world income tax systems involve schedular elements and exempt capital or labor income disproportionately. In Switzerland for instance, interest income accumulates tax free in pension funds and insurance companies, thus distorting the AETR on capital upward relative to countries without such provisions. Carey and Thalmann (2000) find that this bias is substantial, but even correcting for it would leave Switzerland’s AETR on capital well above the OECD average. Moreover, there are other potentially important effects that distort the Swiss AETR on capital less upward than elsewhere; for example, imputed rents of owner occupied housing are taxable in Switzerland, but not, for instance, in the United States. A second problem concerns the choice of the appropriate denominator in the AETR calculations: the gross or the net operating surplus of the economy. The net operating surplus is the correct concept, but economically unfounded differences in depreciation charges might distort the cross-country comparison. Carey and Thalmann (2000) therefore prefer using the gross operating surplus. Switzerland’s AETR on capital calculated on this basis is close to the OECD average, as depreciation charges are particularly large in Switzerland. However, high depreciation charges in Switzerland likely reflect a large capital stock rather than an accounting fiction.

35. The literature on optimal taxation suggests that capital taxation causes particularly large deadweight losses. Indeed, most models suggest that capital income should not be taxed at all, at least if abstracting from equity considerations.14 This is because even small tax rates drive a large wedge between the before-tax and after-tax rate of return in the case of long-term investment. Moreover, real world tax systems typically operate in an environment of positive inflation and tax nominal interest income, thus effectively taxing not only capital income but also capital itself.

36. High taxes on capital income appear not to discourage savings but may be responsible for other distortions. Swiss domestic savings are amongst the highest in the world (Figure II-7). While one could argue in principle that savings would be higher still if it were not for the heavy taxation of capital, the notion that gross domestic savings of close to 30 percent of GDP are suboptimally low remains unappealing.15 More likely, capital does not fully bear the incidence of the tax as it is internationally mobile. In this case saving-consumption decisions remain undistorted. However, the tax system then encourages capital outflows, consistent with Switzerland’s large external current account surplus (Figure II-8), and might give rise to costly avoidance strategies. It would then be more efficient to tax directly those factors that bear the incidence of the tax.

Figure II-7.
Figure II-7.

Selected OECD Countries: Gross Domestic Savings, 1992-2000

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: IMF world Economic outlook.
Figure II-8.
Figure II-8.

Selected OECD Countries: Current Account Balance, 1992-2000

Citation: IMF Staff Country Reports 2002, 107; 10.5089/9781451807240.002.A002

Sources: IMF world Economic outlook.

37. High AETRs on capital income are in any case not inconsistent with high savings if the taxation of savings is non-uniform. AETRs average across taxes on all forms of capital income. Even if the average AETR is high, capital income from some sources might still be taxed very lightly. In this case overall savings might not be discouraged much by taxation of capital, but their allocation might be severely affected. Savings might thus be channeled into unproductive uses with concomitant welfare losses.

38. The efficiency of the Swiss tax system would thus probably improve if the tax burden were rebalanced toward consumption away from capital. Fiscally neutral suggestions in this regard might include phasing-out wealth taxes or providing relief from the current double taxation of distributed profits and capital transaction taxes while compensating the associated revenue loss by an increase of the VAT.16 The VAT rate in Switzerland of 7.6 percent is significantly lower than in other European countries. Rearranging the federal, cantonal, and municipal shares in the VAT could in principle neutralize the distributional implications between levels of governments. Equity concerns could be addressed by increasing progressivity elsewhere in the system, e.g., by removing the cap on income that is subject to unemployment insurance combined with a lowering of the contribution rate.

C. The Structure of Capital Taxation17

39. In Switzerland, capital income taxes vary significantly across assets and investors. At the high-end of the scale is the investment of households in dividend-paying equity: at the corporate level it is subject to corporate income and net wealth taxes and at the personal level it is subject to personal wealth taxes and income taxes on distributed profits. Moreover, the initial investment is made out of taxed income. This sharply contrasts with the tax burden on savings with pension funds: contributions, even the ones over and above the legally required minimum, are tax deductible, and returns accumulate free of capital income and net wealth taxes in the pension funds. Pension benefits are taxed, albeit at a reduced rate if taken out in lump-sum form. Very favorable tax treatment is also accorded to savings with insurance companies with tax free earnings and disbursements, although premiums are not tax deductible. Owner-occupied housing currently attracts an intermediate tax burden: it is subject to property tax, to capital gains tax (in contrast to movable property), and tax on imputed rent, but mortgage interest payments as well as maintenance expenses are tax deductible. Imputed rents tend to be below market rates and capital improvements are in practice often claimed as maintenance.

40. Leveling the playing field between investors is likely to improve efficiency. Tax-induced advantages of certain investors tend to reduce competition between investors and encourage the inefficient use of middlemen. For example, tax advantages provide the domestic insurance and pension fund sectors with room for inefficiencies, which are unlikely to be completely competed away within the sectors. Moreover, they might make it attractive to invest in certain assets through an insurance company or a pension fund even if it were more efficient to do so directly. For example, the tax system might sway a household investor into saving through a pension fund which then invests into real estate that is then rented back to the household investor, even though investing into owner-occupied housing might be less costly overall. The rich asset positions of Swiss pension funds and insurance companies in conjunction with one of the lowest shares of owner-occupied housing in the OECD suggest that this is more than a theoretical possibility.

41. Likewise, levying the playing field between different assets would probably unlock efficiency gains. As argued before, the Swiss tax system favors investment in retained earnings, over loans, over new equity.18 This essentially reflects the absence of a capital gains tax, a marginal corporate income tax rate below the marginal personal income tax rate, and the double taxation of distributed profits. The favorable treatment of retained earnings encourages reinvesting capital in existing firms although the overall most profitable opportunities may lie elsewhere.

References

  • Auerbach, Allan and James Hines Jr. (2001), “Taxation and Economic Efficiency,” NBER Working Paper 8181.

  • Carey, David and Harry Tchilinguirian (2000), “Average Effective Tax Rates on Capital, Labor and Consumption,” OECD Working Paper ECO/WKP(2000)31.

    • Search Google Scholar
    • Export Citation
  • Carey, David, Kathryn Gordon, and Philippe Thalmann (1999), “Tax Reform in Switzerland,” OECD Working Paper ECO/WKP (99)14.

  • IMF (1999), “Switzerland—Selected Issues and Statistical Appendix,” SM/99/31, Chapter III.

  • Queisser, Monika and Dimitri Vittas (2000), “The Swiss Multi-Pillar Pension System: Triumph of Common Sense?,” World Bank, Policy Research Working Paper No. 2416.

    • Search Google Scholar
    • Export Citation
12

Prepared by Christoph Klingen.

13

Note that in the run up to demographic aging a funded pension system does not necessarily spell higher aggregate contributions than a PAYG system. Funded systems need to collect sufficient contributions now to finance the future wave of retirees while PAYG systems delay this until the wave actually retires. But this effect is counterbalanced by the effect of interest earnings by funded pension systems on current contributions. No further adjustment is thus necessary to make the relative revenue takes of Switzerland and Germany comparable.

14

For a recent summary see Auerbach and Hines (2001).

15

Of course one cannot rule out that Swiss savings are higher than elsewhere for completely unrelated reasons. For example, contributions to the funded pension pillar, which tend to, ceteris paribus, increase domestic savings, are much larger that elsewhere. In 2000 they came to 7 percent of GDP. Although only about half of this amount is mandated by law, the remainder also takes on a quasi-mandatory character as employers typically offer their employees only pension plans that go beyond the legally required minimum.

16

Note that current proposals to finance relief from double taxation of distributed profits by introduced a capital gain tax on significant equity holdings would not contribute to rebalancing the tax burden away from capital.

17

This section draws heavily on Carey, Gordon, and Thalmann (1999).

18

See IMF (1999).

Switzerland: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Selected OECD Countries: General Government Tax Revenue, 2000 1/

  • View in gallery

    Selected OECD Countries: General Government Revenue, 2000

  • View in gallery

    Selected OECD Countries: General Government Debt,2001

    (In percent of GDP)

  • View in gallery

    Selected OECD Countries: AETR on Capital, 1991-97

    (in percent)

  • View in gallery

    Selected OECD Countries: AETR on Labor and Consumption Combined, 1991-97

    (In percent)

  • View in gallery

    Selected OECD Countries: Contributions to AETR on Capital by Tax, 1991-1997

  • View in gallery

    Selected OECD Countries: Gross Domestic Savings, 1992-2000

  • View in gallery

    Selected OECD Countries: Current Account Balance, 1992-2000